The World Health Organization (WHO) has endorsed directly observed therapy, short course (DOTS) as the principal strategy for controlling tuberculosis (TB) worldwide. Nevertheless, accumulating evidence suggests that this strategy will be insufficient to produce a reduction in TB incidence. There is also compelling evidence to suggest that isoniazid (INH) preventive therapy, which is not currently recommended by WHO for routine use in developing countries, may be extremely valuable in controlling TB in these countries. While INH preventive therapy is highly efficacious and inexpensive, its use may be limited by poor adherence. Recently completed clinical trials provide evidence supporting the efficacy of shorter, intermittent, chemoprophylactic regiments for preventing TB in patients at high risk for developing active TB. These studies have compared INH and the combination rifampin (Rif) pirazinamide (PZA). In addition, directly observed preventive therapy (DOPT) has been shown to improve completion rates of INH chemoprophylaxis in different settings. Rifapentine (RPT) is a rifamycin- S derivative with antimicrobial activity similar to RIF, but with a longer half-life. Given its efficacy for regimen and the and the available data in animal studies, the efficacy of RPT for the prevention of tuberculosis in people at high risk label trial of short-course, intermittent DOPT. The two arms will be RPT/INH given once weekly for 12 weeks, and Rif/PZA, given twice weekly for 8 weeks. Participants will be recruited from among household contacts of TB cases in the DOTS+arm of Project 1. Randomization will occur at the TB index case level in blocks of 10 (5 in each arm) to assure that an approximately equal number of case contacts are randomized to each arm. The efficacy of each regimen will be determined by comparing the incidence of TB in each arm to the incidence among TB case contacts who receive no prevention therapy in the DOTS arm of Project 1. We hypothesize that both regimens will offer substantial clinical protection against active TB, that RPT/INH will be associated with higher adherence to therapy and comparable rates of toxicity than Rif/PZA. Besides providing, in conjunction with projects 1 and 3, important information for TB control strategies worldwide, this study will provide data on the safety and efficacy of RPT for the prevention of active TB, a drug that, due to its pharmacologic profile, can be used once-weekly, a considerable programmatic advantage.